JPH10334269A - Image processing device and method, and recording medium recording image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform the rendering processing with higher efficiency. SOLUTION: This device has at least a rendering processing part 24 which generates the color data on the pixels to be displayed from the polygon data having the parameters for generation of a polygon ID and its additional position coordinate data and color data. At the same time, a polygon buffer memory 241 is used to store the polygon data together with a frame buffer memory 26 which stores the color data on the pixels included in a frame. Then the part 24 generates Z value from the position coordinate data to show the depth of the pixels included in a polygon within a screen against plural polygons located in the frame. A Z value buffer memory 245 which stores the Z value of pixels included in the frame has a 1st processing part 25 which stores the Z value of the pixel shown on a screen and its corresponding polygon ID and a 2nd processing parts 246 and 247 which generate the color data from the parameter belonging to the polygon ID stored in the memory 245 against each pixel in the frame. The color data on the pixels generated at the parts 246 and 247 are stored in the memory 26.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus using a computer, and more particularly to a method for generating color data of a frame having a plurality of polygons more efficiently in a shorter time or with a less hardware configuration. The present invention relates to an image processing apparatus, a method thereof, and a recording medium on which the image processing program is recorded, which can be performed in a short time.

[0002]

2. Description of the Related Art An image processing technique using a computer is used in a simulation device, a game device, or the like. Normally, polygon data to be drawn on the screen is obtained from image data generated by a simulation or game sequence program, color data for each pixel in the polygon is obtained, and the color data is converted to a frame corresponding to a pixel on the screen. Store in buffer memory. Then, according to the color data in the frame buffer memory, C
An image is displayed on a display device such as an RT.

The above-described processing for obtaining polygon data includes:
Normally, the processing is performed by the geometry processing unit, and the processing of obtaining color data for each pixel from the polygon data is performed by the renderer processing unit.

[0004] The polygon data generated by the geometry processing unit is usually composed of the vertex data. The color data of the pixels in the polygon is obtained by performing an interpolation operation on the parameter values included in the vertex data.

However, there are cases where a plurality of polygons overlap each other in a frame. In this case, only the foremost polygon is displayed on the screen, and other polygons hidden behind one polygon are displayed. Is not displayed. Therefore, conventionally, a Z-value buffer memory corresponding to a pixel in a frame is provided, and when writing color data of a pixel in the frame buffer memory, the Z-value of the pixel is stored in a corresponding pixel area of the Z-value buffer memory. Even writing is done. Then, it is determined whether or not the pixel of the polygon to be processed later is located on the front side of the already written pixel by comparing the respective Z values. Or, as another algorithm,
It is also considered that color data is always written to the frame buffer from the polygon located at the back of the frame.

[0006]

However, when a plurality of polygons overlap each other, the color data of the pixel at the back is once calculated, and after the color data is written in the frame buffer memory, The color data of the pill cell of another polygon on the front side may be obtained by calculation, and the color data may be written to the same address of the frame buffer memory. Therefore, there is a problem that the rendering processing of the pixels on the far side performed earlier is completely wasted, and the efficiency of the rendering processing is deteriorated.

Further, in the conventional method, in the rendering processing of a certain pixel of a certain polygon, the interpolation calculation of the parameter in the vertex data and the interpolation calculation of the Z value are simultaneously performed. Therefore, the hardware configuration of that part becomes very large. Therefore, if a plurality of such parts are to be processed in parallel, the hardware becomes enormous and impractical. This was one of the factors that hindered the efficiency of the rendering process.

It is an object of the present invention to provide an image processing apparatus and method capable of performing image processing using a computer more efficiently, and a recording medium storing the program.

Further, an object of the present invention is to provide an image processing apparatus and method capable of more efficiently performing a rendering process for generating pixel color data in image processing, and a recording medium on which the program is recorded. It is in.

[0010]

According to the present invention, the object of the present invention is to provide a method for converting color data of a pixel to be displayed from polygon data having at least a polygon ID and position coordinate data and a parameter for generating color data. An image processing apparatus having a rendering processing unit for generating the polygon data, comprising: a polygon buffer memory for storing the polygon data; and a frame buffer memory for storing color data of pixels in a frame, wherein the rendering processing unit For multiple polygons located within
A Z value indicating the depth of the pixel in each polygon in the screen is generated from the position coordinate data, and is displayed on the screen in a Z value buffer memory in which the Z value of the pixel in the frame is stored. A first processing unit for storing a Z value of a pixel and the corresponding polygon ID, and for each pixel in the frame, color data from the parameter belonging to the polygon ID stored in the Z value buffer memory; A second processing unit for generating the image data, wherein the color data of each pixel generated by the second processing unit is stored in the frame buffer memory. Is done.

With this configuration, the Z value of a pixel is first determined for all polygons in a frame,
Pixels to be displayed on the screen are determined according to the Z value, and thereafter, color data can be generated for the pixels to be displayed in the frame. Therefore, it is possible to avoid unnecessary generation of the color data of the pixels in the overlapping portion.

Further, the present invention is characterized in that the first processing unit is provided in a plurality of layers, and Z values are generated for a plurality of pixels in parallel. Also, the first processing unit is provided in a plurality of layers, and the Z of a pixel is determined for a plurality of polygons.
It is characterized in that value generation is performed in parallel.

Further, according to the present invention, there is provided a rendering method for generating color data of a pixel to be displayed from polygon data having at least a polygon ID and position coordinate data and a parameter for generating color data. In the image processing method having a processing step, the rendering processing step generates, from the position coordinate data, a Z value indicating a depth in a screen of a pixel in each polygon for a plurality of polygons positioned in the frame. A first processing step of storing a Z value of a pixel displayed on the screen and the corresponding polygon ID in a Z value buffer memory in which a Z value of a pixel in the frame is stored; For each pixel in the pixel, a color is determined from the parameter belonging to the polygon ID stored in the Z-value buffer memory. It generates over data, is achieved by providing an image processing method characterized by having a second processing step of storing the color data of each pixel thus generated in the frame buffer memory.

Further, according to the present invention, there is provided a computer having at least a central processing unit for performing arithmetic processing and a frame buffer memory for storing color data of pixels in a frame. A computer-readable recording medium storing a program for executing an image processing procedure including a rendering process for generating color data of a pixel to be displayed from polygon data having position coordinate data and parameters for generating color data The program generates, for the plurality of polygons located in the frame, a Z value indicating the depth of the pixels in each polygon on the screen from the position coordinate data, and calculates the Z value of the pixel in the frame. In the stored Z value buffer memory,
A first processing procedure for storing a Z value of a pixel displayed on the screen and the corresponding polygon ID, and for each pixel in the frame, a polygon ID stored in the Z value buffer memory. A second processing procedure for generating color data from the parameters belonging thereto and storing the generated color data of each pixel in the frame buffer memory, and a program for causing a computer to execute the recording. Achieved by providing a medium.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. However, the technical scope of the present invention is not limited to the embodiment.

FIG. 1 is a diagram showing a case in which two polygons ID0 and ID1 partially overlapping in a frame (screen) 10 are displayed. In this example, polygon ID0 is composed of vertices 00, 01, and 02, and polygon ID1 is
Vertices 10, 11, and 12 are formed. Then, in the screen, the polygon ID0 is located ahead (front) than the polygon ID1. Therefore, the hatched portion 12 in the figure
Area, two polygons overlap each other.

However, as described above, in the area 12, only the foremost polygon ID0 is displayed.
The polygon ID1 located behind it is not displayed.
Therefore, if the color data of the pixel in the polygon ID 1 located in the area 12 is first obtained by the interpolation calculation and written in the frame buffer memory, the interpolation calculation and the writing to the memory are useless processing. Becomes That is,
This is because the color data of the pixel of the polygon ID0 obtained thereafter is overwritten again in the frame buffer memory 26.

The polygon data generated by the geometry processing unit based on image data from a computer is generally generated in an at random order. Therefore, as described above, the case where the data of the polygon ID1 is generated first and the rendering process is performed occurs simply with a probability of 50%.

FIG. 2 is a diagram showing an example of the configuration of polygon data generated by the geometry processing unit. In this example, polygon ID 0 is composed of vertices 00, 01, and 02, and the data of each vertex includes a plurality of vertex parameters. In the example of FIG. 2, the vertex parameters are the screen coordinates (Sx, Sy) of the vertex on the screen, the Z value (depth value) indicating the depth in the screen, and the texture coordinates (Tx, Tx, Ty), a normal vector (Nx, Ny, Nz) and an α value (scalar value) representing transparency. These vertex parameters are given as attribute data for each vertex. Also, the polygon I
For D, the start address of each memory when data of each polygon is stored in memories 201 and 202 (FIG. 4) to be described later can be used.

FIG. 3 is a diagram for explaining an example in which each parameter of a pixel in polygon ID 0 is obtained by interpolation using vertex parameters. As shown in FIG. 3A, X and Y coordinates are defined, and an interpolation operation is performed on a pixel G in the polygon from the vertex 00 in the scanning direction indicated by the arrow. This scanning direction is an example, and is not necessarily limited to the case of moving in parallel to the X axis and the Y axis, but is a scanning direction defined by an algorithm when scanning in a direction inclined at a predetermined angle from that. May be.

In the example shown in FIG.
Moves from the vertex 00 as a starting point in the positive direction of the Y axis, moves along the horizontal line between the left and right ridge lines ab and ac in the positive direction of the X axis, and further moves in the positive direction of the Y axis. To move between the left and right ridges of the horizontal line. Then, in this example, the vertex 02 is the end point.

FIG. 3B shows the internal division ratios t0, t1, and t2 used in the interpolation operation. Now, it is assumed that the interpolation calculation of the parameter for the pixel G is performed. To determine texture coordinates, which is one of the parameters, vertex 00,
The texture coordinates of 01 and 02 are (Tx0, Ty0) (Tx1,
Ty1) (Tx2, Ty2), the texture coordinates (Txd, Tyd) of the point d on the ridge line ab are as follows: Txd = Tx0.t0 + Tx1. (1-t0) Tyd = Ty0.t0 + Ty1. (1-t0) Similarly, the texture coordinates (T
xe, Tye) is as follows: Txe = Tx0 · t1 + Tx2 · (1−t1) Tye = Ty0 · t1 + Ty2 · (1−t1) Therefore, the texture coordinates (Txg, Tyg) of the pixel G on the horizontal line de are as follows: Txg = Txd.t2 + Txe. (1-t2) Tyg = Tyd.t2 + Tye. (1-t2)

Screen coordinates of other vertex parameters and Z
The value, the normal vector, the α value, and the like are obtained by the same interpolation calculation. Each time the pixel G is moved, the internal division ratio t0,
t1 and t2 are incremented. As described above, the calculation for obtaining the pixel parameters by the interpolation method requires a great deal of computer processing time.

[Example of Image Processing Apparatus] FIG. 4 is a schematic block diagram of an image processing apparatus according to an embodiment of the present invention. The block diagram includes a CPU 20 that constitutes a computer that generates image data by executing a game sequence program or the like, and performs polygon conversion and other operations in accordance with the image data from the CPU 20 as shown in FIG. A geometry processing unit 22 that generates
It comprises a renderer processing unit 24 that generates color data for each pixel based on the polygon data. 26 is a frame buffer memory for storing the color data, and 28 is a display device for displaying the color data.

In the computer, a RAM 201, a ROM 202, and an input / output device I / O which are usually connected to the CPU 20 are provided.
O203 is provided. The above-mentioned sequence program is stored in the ROM 202, for example.

The geometry processing unit 22 performs a geometry conversion for changing the position of the polygon, a clipping process for extracting a polygon in the screen according to the viewpoint data, and a perspective conversion for obtaining two-dimensional coordinates on the screen from three-dimensional coordinates. Generally done. Then, the screen coordinates and the Z value (depth value) indicating the depth in the screen are given as one of the vertex parameters as the position data of the vertex.

The renderer processing unit 24 performs an interpolation operation by interpolating the Z values of the points d and e on the ridge line (edge) of the polygon by the internal division ratios t0 and 1, respectively. Edge interpolator 242,
A raster interpolator 243 and a Z comparator 244 for interpolating the Z value of the pixel G in the horizontal line de at the internal division ratio t2, a Z value buffer memory 2 storing the Z value, the internal division ratio, and the polygon ID.
45.

The edge interpolator 242 and the raster interpolator 2
43, Z value comparator 244 and Z value buffer memory 245
Is a unit 2 for finding the Z values of all the pixels in the polygon
5 is constituted.

Further, the renderer processing unit 24 calculates the parameters of each pixel in the frame according to the internal division ratio and the polygon ID stored in the Z-value buffer memory 245, based on the parameters obtained by the interpolation. Texture generation unit 2 for obtaining color data of each pixel
47. Reference numeral 248 denotes a texture map memory storing texture data.

The texture generator 247 described above calculates the texture coordinates (Txg, Txg,
Tyg), texture data in the texture map memory 248 is read, and normal vectors (Nxg, Nyg,
Nzg) to perform a shading process (shading process), and use the transparency αg to perform a process of blending colors (blending) into translucent pixels. Further, the parameter interpolator 246 is provided in the polygon buffer memory 24.
1 is also connected to reference the polygon data stored therein.

One example of a characteristic part of the configuration of the above-described image processing apparatus is that an edge interpolator 242, a raster interpolator 243, and a parameter interpolator 246 for performing an interpolation operation of the Z value of a pixel are provided. It is a point that is separated. That is, for all polygons included in a certain frame, the Z value is first obtained by interpolation, and the Z value buffer memory 2
45, the polygon ID to which the pixel located closest in the screen belongs, and the internal division ratios t0 and t of the polygon
1 and t2 are stored. After that, the Z value buffer memory 24
The parameter interpolator 246 performs a parameter interpolation operation only on the pixels stored in 5. Therefore, the texture generation unit 247 only performs the calculation of generating the color data of the displayed pixel, and only writes the color data to the displayed pixel in the frame buffer 26. Therefore, the generation calculation of the color data of the area 12 of the polygon ID1 located behind shown in FIG. 1 and the writing to the frame buffer memory 26 are avoided.

[Image Processing Flow] FIG. 5 is a flowchart of image processing by the above-described image processing apparatus. FIG.
An example of the image processing method will be described below in detail with the block diagram of FIG.

First, image data is generated by the CPU 20 (S1). The image data includes polygon movement data and viewpoint data. The geometry processing unit 22 performs the above-described processes such as the geometry conversion, the clipping process, and the perspective conversion, and generates polygon data as shown in FIG. 2 and stores it in the polygon buffer memory 241 (S2). The polygon data is generated one after another, for example, at random from the geometry processing unit 22. Then, rendering processing is performed on all the polygon data in one frame. Therefore, for example, by providing the polygon buffer memory 241 for two frames, while the rendering process is performed on the polygon in one frame, the polygon data in the next frame can be stored in the other polygon buffer memory. .

On the basis of the Z value of the vertex of a certain polygon data stored in the polygon buffer memory 241, the Z value of the pixel in the polygon is interpolated. First, the Z value of the vertex is interpolated and the left edge (edge)
The Z value of point d on ab is determined (S3). In this interpolation calculation, the internal division ratio t0 is used. Similarly, the z value of the point e on the right ridge line (edge) ac is determined (S4). In this interpolation calculation, the internal division ratio t1 is used. These interpolation operations are
This is performed in the edge interpolator 242. Thereafter, the Z value of the pixel G is obtained by performing raster interpolation based on the Z values of the points d and e on both sides (S5). In this interpolation calculation, the internal division ratio t2 is used.

4 and 5, as an example of the Z value,
1 / z is shown because it is convenient to use the reciprocal of the Z value in order to interpolate after being perspectively transformed and displayed on the display screen.

The Z value of the pixel G obtained in the above steps S3, S4, S5 is compared with the Z value already written in the Z value buffer memory 245 by the Z value comparator 244, and the Z value in the near side of the screen is displayed. Is determined (S6). When the pixel being processed is closer to the front (when the Z value is small or the 1 / z value is large), the Z value of the pixel being processed, the polygon ID, and the internal division ratios t0, t1, t2 are set. Is overwritten. Therefore, in the Z value buffer memory 245, the Z value of the foremost pixel in the screen, the corresponding polygon ID data, and the internal division ratios t0, t1, and t2 used for the interpolation calculation of the pixel are stored. Is done.

As shown in FIG. 5, steps S5, S6, S7
Is repeated by the size of the horizontal line de shown in FIG. 3 in the X-axis direction. Then, when the raster scan of the horizontal line de ends, the horizontal line de is shifted in the positive direction of the Y axis, and the processing of steps S3, 4, 5, 6, and 7 is performed again on the new horizontal line. Therefore, steps S3, 4, 5, 6, and 7 are repeated for the size in the Y-axis direction shown in FIG.

As described above, when the interpolation of the Z value for the pixels in one polygon and the writing to the buffer memory 245 by the Z value comparison are completed, the process proceeds to step S3 for the next polygon. , 4, 5, 6, and 7 are performed. As shown in FIG. 5, interpolation of Z values is performed for all polygons in the frame,
The Z value of the pixel located at the foreground in the screen, the polygon ID data to which the pixel belongs, and its internal division ratio are stored in the Z value buffer memory 2.
45.

In the above calculation, the Z value interpolation calculation may be useless for pixels located at the back of the screen.
However, the Z-value interpolation calculation can be performed relatively easily, and rarely causes a significant decrease in the efficiency of the entire image processing. Further, the internal division ratio of the pixel obtained by the interpolation calculation of the Z value is stored in the buffer memory 245, and is used for the interpolation calculation of the subsequent parameters.

Now, after the polygon ID data of the pixel to be displayed on the display screen and its internal division ratio are stored in the Z-value buffer memory, the parameter interpolator 24
6 and the texture generation unit 247 generate pixel color data. First, the Z value of a pixel from the Z value buffer memory 245 is
The values, internal division ratios t0, t1, t2 and polygon ID data are read (S8). Then, texture coordinates, α value, and normal vector, which are vertex parameters of the polygon ID, are read from the polygon buffer memory 241 (S
9). Internal division ratios t0, t for those vertex parameters
The pixel parameters are interpolated using 1, 1 and t2 (S10).

Then, based on the obtained parameter values,
The color data of the pixel is generated by the texture generation unit 247 (S11). Specifically, texture coordinates (Txg, T
yg), the texture data in the texture map 248 is read. Then, the normal vectors (Nxg, Nyg, N
Based on zg), a shading process operation is performed on the color data of the pixel to generate color data subjected to shading. If it is determined from the α value that the pixel is a translucent pixel, the pixel is blended with the color data of the pixel on the back side (blending processing).

Since the blending process is not directly related to the characteristic portion of the present invention, it will not be described in detail. For example, data on whether or not a translucent polygon is provided as attribute data of the polygon is supplied from the computer. This can be realized by rendering the translucent polygon after rendering the opaque polygon.

Then, the color data of the pixel obtained as described above is written in the frame buffer memory 26 (S
12).

The processes from steps S8 to S12 are repeated for the number of pixels in the buffer memories 245 and 26. Since the processing here is performed only for the number of pixels to be displayed, it can be performed in a very short processing time. In particular, since the processing in the texture generation unit 246 requires a relatively long processing time, reducing the number of pixels to be processed in the texture generation unit 246 requires:
This contributes to improving the efficiency of the entire rendering process. In addition, since the internal division ratio of the pixel already obtained at the time of the Z value interpolation calculation can be used, the processing time can be shortened in that respect as well.

[Example of Image Processing Apparatus by General-Purpose Computer] The image processing apparatus shown in FIG.
2 and the rendering processing unit 24 are configured by dedicated hardware. In the case of this example, the image data is substantially sequentially converted into polygon data and pixel color data by pipeline processing.

However, in the image processing method of the present invention, it is possible to realize the above-described geometry processing and rendering processing by a software program using a general-purpose computer. The image processing program in that case has a program code for causing a computer to realize each processing of the processing flowchart shown in FIG.

FIG. 6 is a configuration diagram of a computer when the image processing method of the present invention is realized using a general-purpose computer. In this example, a CPU 30, a RAM 31, a program memory 32, an I / O device 33, a polygon buffer memory 34, a Z value buffer memory 35, and a frame buffer memory 36 are connected via a common bus 38. An external display device 37 is connected to the frame buffer memory 36.

The program memory 33 includes, for example, a magnetic recording medium such as a hard disk device, a recording medium on which writing and reading are performed magneto-optically, a CDROM, a semiconductor memory, and the like. In the program memory 33, a game or simulation program and an image processing program are recorded. RAM 31 is a CP
It is used as a work memory used for various arithmetic processing of U. Therefore, the polygon buffer memory 3
4 and the Z-value buffer memory 35 are connected to a high-speed accessible R
It may be formed in the AM.

When the image processing process of the present invention is realized using a general-purpose computer, the program memory 3
The above-described image processing is realized by a computer using the program code of the image processing program stored in the computer 2. Therefore, the program memory needs to be a computer-readable recording medium.

[Other Embodiments] FIG. 7 is a block diagram showing another configuration example of the image processing apparatus of the present invention. This example is not a general-purpose computer but an example of a hardware configuration that performs processing by a pipeline method as shown in FIG.

In this configuration example, the Z-value buffer generation unit 25 shown in FIG. 4 has a plurality of units, and a Z-value buffer is generated in parallel for a plurality of pixels.
Specifically, as shown in FIG. 7, four units 25A, 25B, 25C, and 25D for generating a Z-value buffer are provided in parallel. Therefore, the Z value buffer memories 245A, 245
B, 245C and 246D are divided into four. The latter half of the rendering processing unit such as the parameter interpolator 245 and the texture generation unit 247 is provided only once as in the case of FIG.

FIGS. 8 and 9 are diagrams for explaining an image processing method by the image processing apparatus of FIG. The diagram shown in FIG. 8 explains that the Z value buffer memory is divided into four. In this example, the pixels in the frame 10 are denoted by 1,
Divide as shown in 2, 3, and 4. Then, a Z-value buffer memory 245A is provided for the pixel of number 1.
Similarly, a Z value buffer memory 245B is provided for the pixel of No. 2, and Z value buffer memories 245C and 245D are provided for the pixels of No. 3 and 4, respectively. Therefore, each Z value buffer memory stores the image of frame 10 in 4
The Z value of the image thinned to one-half is stored.

The process of interpolating the Z value for a certain polygon and storing the Z value, polygon ID data, and internal division ratio in the Z value buffer memory is performed in parallel with the pixels of numbers 1 to 4. Will be That is, as shown in FIG. 9, the interpolation operation of the Z value is performed in parallel on the pixels G1, G2, G3, and G4 by four units.

For pixel G1, unit 25A
The edge interpolator 242A of the inside calculates the internal division ratios t01 and t11, and obtains the Z value at the points d1 and e1 on the ridge lines ab and ac by interpolation. Then, the raster interpolator 243A obtains the internal division ratio t21 and calculates the Z of the pixel G1 on the horizontal line d1e1.
Find the value.

For the pixel G2, the unit 25B
The edge interpolator 242B (shown hidden from the drawing) in
Find internal division ratios t01 and t11, and point d on ridgelines ab and ac
The Z value at 1, e1 is obtained by interpolation. Then, the raster interpolator 243B obtains the internal division ratio t22 and obtains the horizontal line d1.
Find the Z value of pixel G2 on e1. Therefore, since the edge interpolator performs the same processing for the pixels G1 and G2, they can be shared. However, in the examples of FIGS. 8 and 9, since the scanning of the pixels is parallel to the Y axis, only the internal division ratio of the edge interpolation of the pixels G1 and G2 is common. Therefore, when the pixels G1 and G2 are inclined with respect to the Y axis or when scanning is performed by another algorithm, the internal division ratio of the edge is different.

For the pixels G3 and G4, the interpolation of the Z value is performed by the units 25C and 25D in the same manner as described above.

The above-described parallel processing simply performs steps S3 to S7 of the flowchart shown in FIG. 5 in parallel.
Therefore, when the interpolation of the Z value for all the polygons in the frame and the writing to the Z value buffer memory are completed, the polygon ID data stored in the four Z value buffer memories 245A, 245C, and 245D and the internal division ratio , The interpolation calculation of the parameters of each pixel and the generation processing of the color data using the obtained parameters are performed. The parameter interpolation calculation is performed by the interpolator 246, and the color data generation processing is performed by the texture generation unit 247. This is the same as in FIG.

FIG. 10 is a block diagram showing another example of the configuration of the image processing apparatus according to the present invention. FIG. 11 is a diagram for explaining the image processing method. In this example,
Four Z value generation units 27A, 27B, 27C, 27
A common edge interpolator 242 is provided for D. Each unit has a raster interpolator 243A,
The Z value comparator 244A and the Z value buffer memory 245A
Each is provided.

Referring to FIG. 11, the edge interpolator 2
It is understood that 42 is common. As shown in FIG. 11, four pixels G1, G2, G processed in parallel
3 and G4 are adjacent on the same horizontal line de. Therefore,
The operation of the edge interpolator is the same. Therefore, the edge interpolator 242 obtains the Z values of the points d and e on the ridge line using the internal division ratios t0 and t1. And four units 27
According to A, B, C, and D, the internal division ratios t21, t22, and t2
3, t24, each pixel G1, 2,
Find Z values of 3 and 4.

With the configuration shown in FIG. 10, the hardware configuration of the four units can be further reduced.
It is possible to minimize an increase in cost due to providing two units in duplicate.

FIG. 12 is a block diagram showing another example of the configuration of the image processing apparatus according to the present invention. In this example, the configuration is such that the Z-value interpolation calculation of a plurality of polygons is processed in parallel. Therefore, the Z value buffer memory 245 is provided commonly for the Z value generation units 29A, B, C, and D. Each unit has an edge interpolator 2
42A-D, raster interpolators 243A-D, Z value comparator 2
44A to 44D are provided. Then, for the four polygons, the units 29A to 29D perform the Z value interpolation calculation in parallel. Since the Z-value buffer memory 245 is commonly accessed from each unit, common access is performed by, for example, allocating an access time in a time-sharing manner.

Further, as another configuration example, a configuration is adopted in which a plurality of polygons are processed in parallel as shown in FIG. 12, and a plurality of pixels can be processed in parallel in the unit for each polygon as shown in FIG. 7 and FIG. The configuration may be any.
Therefore, if four units are provided for each polygon, the Z-value buffer memory also needs to be divided and provided as shown in FIGS. The best combination is selected from the hardware scale and processing efficiency.

In the above description of the embodiment, an example has been described in which pixels in a polygon are scanned in a direction parallel to the X axis and the Y axis. However, the invention is not limited to such a scanning algorithm. For example, the present invention can be applied to scanning in a direction inclined by a predetermined angle from the X axis and the Y axis. Alternatively, the pixels may be scanned in accordance with the complex coordinates including the angle and the length. Furthermore, the present invention can also be applied to a case where scanning is performed by an algorithm in which the center of the polygon is set as the origin, the center of the area divided into four quadrants, and the center of the four quadrant area with respect to the center.

In the above example, the screen coordinates (Sx, Sy) of the vertices are used as an example of the position coordinate data of the polygon.
Although the description has been made with reference to the example of the Z-value and the Z value, another example may be the world coordinates (x, y, z). This is because the only difference is whether the perspective transformation is performed before or after writing to the buffer.

[0065]

As described above, according to the present invention, the rendering processing for obtaining the color data of the pixels in a plurality of polygons can be performed more efficiently. That is, the first half of Z-value comparison and storage in the Z-value buffer memory from the Z-value interpolation calculation by the rendering processing unit, and parameter interpolation and color data generation using polygon ID data in the Z-value buffer memory. Doing the latter part is separated. Then, in the first half, processing is performed on all the polygons in the frame to determine the pixels to be displayed on the screen. Thereafter, color data is generated for those pixels. Therefore, when there are overlapping polygons, wasteful generation of color data can be avoided. As a result, the overall rendering processing efficiency is improved. Moreover, the Z-value interpolation calculation itself is a simpler process than the generation of color data. Even if the Z-value interpolation calculation is performed on overlapping polygons, it is necessary to reduce processing efficiency so much. No.

Further, by providing a plurality of hardware configurations in the first half, it is possible to perform Z value generation processing for a plurality of pixels or a plurality of polygons in parallel.
In addition, since the first half and the second half are separated, the hardware configuration of the first half is smaller than that of the entire renling processing unit. Therefore, even if hardware is provided redundantly for parallel processing, the hardware configuration is not so large.

[Brief description of the drawings]

FIG. 1 is a diagram showing a case where two polygons ID0 and ID1 partially overlapping in a frame (screen) 10 are displayed.

FIG. 2 is a diagram illustrating a configuration example of polygon data generated by a geometry processing unit.

FIG. 3 is a diagram illustrating an example of obtaining each parameter of a pixel in a polygon by an interpolation method using vertex parameters.

FIG. 4 is a schematic block diagram of an image processing apparatus according to an embodiment of the present invention.

FIG. 5 is a flowchart of image processing performed by the image processing apparatus.

FIG. 6 is a configuration diagram of a computer when the image processing method of the present invention is realized using a general-purpose computer.

FIG. 7 is a block diagram illustrating another configuration example of the image processing apparatus of the present invention.

FIG. 8 is a diagram for explaining an image processing method by the image processing device in FIG. 7;

9 is a diagram for explaining an image processing method by the image processing device of FIG. 7;

FIG. 10 is a block diagram showing still another configuration example of the image processing apparatus of the present invention.

FIG. 11 is a diagram for explaining the image processing method of FIG. 10;

FIG. 12 is a block diagram showing still another configuration example of the image processing apparatus of the present invention.

[Explanation of symbols]

24 rendering processing unit 25 first rendering processing unit 246, 247 second rendering processing unit (parameter interpolator, texture generation unit) 241 polygon buffer memory 242, 243 Z value interpolator 244 Z value comparator 245 Z value buffer memory 248 Texture Map 26 Frame Buffer Memory

Claims (29)

[Claims] 1. An image processing apparatus having a rendering processing unit for generating color data of a pixel to be displayed from polygon data having at least a polygon ID and position coordinate data belonging to the polygon ID and a parameter for generating color data. A polygon buffer memory for storing data, and a frame buffer memory for storing color data of pixels in a frame, wherein the rendering processing unit performs, for a plurality of polygons located in the frame, A Z value indicating the depth of the pixel in the screen is generated from the position coordinate data, and the Z value of the pixel displayed on the screen is stored in the Z value buffer memory in which the Z value of the pixel in the frame is stored. A first processing unit for storing the polygon ID corresponding thereto; A second processing unit for generating, for each pixel, color data from the parameter belonging to the polygon ID stored in the Z-value buffer memory, wherein the color of each pixel generated by the second processing unit is provided. An image processing apparatus, wherein data is stored in the frame buffer memory. 2. The image processing apparatus according to claim 1, wherein the first processing unit performs a Z-value interpolator that performs a Z-value interpolation operation on the pixel based on the position coordinate data. And a Z-value comparator for determining whether or not the screen is located in front of the screen based on the Z-value obtained by the device and the Z-value in the Z-value buffer memory. 3. The image processing apparatus according to claim 2, wherein the Z value interpolator obtains an internal division ratio of the pixel to be used for the interpolation operation, and stores the Z value in the Z value buffer memory together with the Z value. The internal division ratio is stored. 4. The image processing apparatus according to claim 3, wherein the second processing unit determines a parameter of the pixel based on a parameter of the polygon data and an internal division ratio stored in the Z value buffer memory. It is characterized by having a parameter interpolator for performing an interpolation operation on a value. 5. The image processing apparatus according to claim 1, wherein the second processing unit performs an interpolation operation on a parameter value of the pixel based on a parameter of the polygon data, and the parameter interpolator. And a color data generation unit that generates the color data based on the parameter value of the pixel generated by the color data generation unit. 6. The image processing apparatus according to claim 1, wherein the parameters of the polygon include at least texture coordinates, and the second processing unit obtains the texture coordinates of the pixel by interpolating the texture coordinates. Reading the texture data stored at the texture coordinates of the pixel from the texture map storing the texture data of the polygon to generate the color data. 7. The image processing apparatus according to claim 6, wherein the parameters of the polygon further include a normal vector, and the second processing unit interpolates the normal vector to obtain a normal of the pixel. A vector is obtained, and shading processing is performed on the obtained color data according to the normal vector. 8. The image processing apparatus according to claim 1, wherein the polygon data includes a vertex ID, the position coordinate data of a vertex belonging to the vertex ID, and the parameter. 9. An image processing apparatus having a rendering processing unit for generating color data of a pixel to be displayed from polygon data having at least a polygon ID, position coordinate data belonging to the polygon ID, and a parameter for generating color data. A polygon buffer memory for storing data, and a frame buffer memory for storing color data of pixels in a frame, wherein the rendering processing unit performs, for a plurality of polygons located in the frame, A Z value indicating the depth of the pixel in the screen is generated from the position coordinate data, and the Z value of the pixel displayed on the screen is stored in the Z value buffer memory in which the Z value of the pixel in the frame is stored. A first processing unit for storing the polygon ID corresponding thereto; A second processing unit for generating color data from the parameters belonging to the polygon ID stored in the Z-value buffer memory for each pixel, wherein the color of each pixel generated by the second processing unit is Data is stored in the frame buffer memory; the first processing unit is provided in a plurality of layers;
An image processing apparatus, wherein values are generated in parallel. 10. The image processing apparatus according to claim 9, wherein the first processing unit performs an interpolation calculation of a Z value of a pixel based on the position coordinate data, and the Z value interpolator. And a Z-value comparator for judging whether or not the screen is located in front of the screen based on the Z-value obtained by the above and the Z-value in the Z-value buffer memory. Characteristically, value generation is performed in parallel. 11. The image processing apparatus according to claim 10, wherein said Z value interpolator generates an Z value of a point on an edge of the polygon and a Z value on an edge generated by said edge interpolator. A raster interpolator that performs an interpolation operation on the Z value of the pixel between the edges according to the value. 12. The image processing apparatus according to claim 11, wherein said edge interpolator is provided commonly to a plurality of first processing units. 13. The image processing apparatus according to claim 9, wherein the first processing unit performs an interpolation operation on a Z value of a pixel based on the position coordinate data, and the Z value interpolator. Has a Z-value comparator for determining whether or not the screen is located in front of the screen based on the Z-value obtained in the Z-value buffer memory and the Z-value in the Z-value buffer memory. The method is characterized in that the generation of the Z value of the pixel is performed in parallel. 14. An image processing method comprising: a rendering processing step of generating color data of a pixel to be displayed from polygon data having at least a polygon ID, position coordinate data belonging thereto, and parameters for generating color data. In the processing step, for a plurality of polygons located in the frame, a Z value indicating the depth of the pixels in each polygon in the screen is generated from the position coordinate data, and the Z value of the pixels in the frame is stored. A first processing step of storing a Z value of a pixel displayed on the screen and the corresponding polygon ID in a Z value buffer memory to be processed; and a Z value buffer for each pixel in the frame. Color data is generated from the parameters belonging to the polygon ID stored in the memory, and each of the generated An image processing method characterized by having a second processing step of storing the color data of the cell in the frame buffer memory. 15. The image processing method according to claim 14, wherein said first processing step includes a Z-value interpolation step of performing a Z-value interpolation operation of said pixel based on said position coordinate data. A Z value comparing step of determining whether or not the screen is located in front of the screen based on the Z value obtained in the step and the Z value in the Z value buffer memory. 16. The image processing method according to claim 15, wherein the Z value interpolation step further includes a step of calculating an internal division ratio of the pixel to be used for the interpolation operation, wherein the first processing is performed. The step further includes a step of storing the internal division ratio together with the Z value in the Z value buffer memory. 17. The image processing method according to claim 16, wherein said second processing step comprises: setting a parameter of said pixel based on a parameter of said polygon data and an internal division ratio stored in said Z value buffer memory. It is characterized by having a parameter interpolation step of performing an interpolation operation on the value. 18. The image processing method according to claim 14, wherein the second processing step includes: a parameter interpolation step of performing an interpolation operation on a parameter value of the pixel based on a parameter of the polygon data; And a color data generating step of generating the color data based on the parameter values of the pixel generated in (1). 19. The image processing method according to claim 14, wherein the parameters of the polygon include at least texture coordinates, and the second processing step determines the texture coordinates of the pixel by interpolating the texture coordinates. Reading the texture data stored at the texture coordinates of the pixel from the texture map storing the texture data of the polygon to generate the color data. 20. The image processing method according to claim 14, wherein the first processing step is performed on a plurality of pixels in parallel. 21. The image processing method according to claim 14, wherein the first processing step is performed on a plurality of polygons in parallel. 22. A computer having at least a central processing unit for performing arithmetic processing and a frame buffer memory for storing color data of pixels in a frame is provided with at least a polygon ID, position coordinate data belonging thereto, and parameters for generating color data. From the polygon data with
In a computer-readable recording medium on which a program for executing an image processing procedure having a rendering process for generating color data of a pixel to be displayed is recorded, the program is provided for each of a plurality of polygons located in a frame. A Z value indicating the depth of a pixel in the polygon in the screen is generated from the position coordinate data, and a Z value buffer memory in which the Z value of the pixel in the frame is stored is stored in a Z value buffer memory. A first processing procedure for storing the Z value of the polygon and the corresponding polygon ID, and for each pixel in the frame, generating color data from the parameter belonging to the polygon ID stored in the Z value buffer memory And storing the color data of each of the generated pixels in the frame buffer memory. And management procedures, recording medium characterized by having a program for causing a computer to execute. 23. The recording medium according to claim 22, wherein the first processing procedure includes: a Z-value interpolation procedure for performing a Z-value interpolation operation of the pixel based on the position coordinate data; And a Z-value comparison procedure for determining whether or not the screen is located in front of the screen based on the Z-value obtained in (1) and the Z-value in the Z-value buffer memory. 24. The recording medium according to claim 23, wherein the Z-value interpolation step further includes a step of obtaining an internal division ratio of the pixel to be used for the interpolation operation, wherein the first processing step is performed. The method further comprises a step of storing the internal division ratio together with the Z value in the Z value buffer memory. 25. The recording medium according to claim 24, wherein the second processing procedure is performed based on a parameter value of the pixel based on a parameter of the polygon data and an internal division ratio stored in the Z value buffer memory. Is characterized by having a parameter interpolation procedure for performing an interpolation calculation of 26. The image processing method according to claim 22, wherein the second processing procedure includes: a parameter interpolation procedure for performing an interpolation operation on a parameter value of the pixel based on a parameter of the polygon data; And a color data generation procedure for generating the color data based on the parameter value of the pixel generated in (1). 27. The recording medium according to claim 22, wherein the parameters of the polygon include at least texture coordinates, and the second processing step interpolates the texture coordinates to obtain texture coordinates of the pixel. Reading the texture data stored in the texture coordinates of the pixel from the texture map storing the texture data of the polygon, and generating the color data. 28. The recording medium according to claim 22, wherein said first processing procedure is performed on a plurality of pixels in parallel. 29. A recording medium according to claim 22, wherein said first processing procedure is performed in parallel on a plurality of polygons.